DESCRIPTION: Regulation of the MUC2 locus and the consequences of loss of mucins in colon tumorigenesis will be investigated by pursuing the effects of an Apc mutation on goblet cell differentiation and the ability of loss of mucins to promote tumorigenesis in genetically initiated tumors. Novel genetic mouse models will be used in which there is targeted inactivation of the mouse Muc2 gene, and goblet cells are specifically labelled by the expression of beta-gal linked to the promoter of the human MUC2 gene. The project stems from three observations: (1) In human aberrant crypt foci (ACF) early preneoplastic lesions, there is depletion of goblet cells and a depletion of the mucins they produce. (2) The genetically initiated mouse model Apc 1638, which has a targeted inactivation of the Apc gene and recapitulates the phenotype of FAP individuals, also develops ACF that, as in humans, show depletion of goblet cells. (3) The phenotype of a genetically initiated mouse (i.e., with a mutation in apc) is attenuated by cyclooxygenase 2 (COX2) gene inactivation, providing compelling evidence that COX2 is involved in the development of colon cancer. The cloning, structural and functional analysis of the mouse MUC2 locus will be completed for two reasons: to guide construction of transgenics to identify the elements necessary for lineage specific expression (to be used under Aim 4) by comparison of the human and mouse regulatory elements; and to provide reagents for development of the Muc2 null mouse. Since a loss of mucins is an early event in ACF formation in human and mouse models, the Muc2-/- mouse to be developed, in combination with the Apc 1638 mouse (GI tumorigenesis is preceded by the development of ACF), will permit the determination of how mucin loss influences tumor development. The MUC2/beta-gal transgenic mice will also allow determination of whether mutation of Apc perturbs the differentiation pathway of the goblet cell lineage; alterations in the timing and early pathological lesions of this process will also be defined. Finally, studies will test the hypothesis that the loss of protective mucins (which happens in ACF, early in progression) leads to an inflammatory response that may be linked to induction of COX2 (a critical element in colon tumorigenesis), thereby demonstrating that loss of mucins can be a promoter of APC-initiated tumors. Mouse models already available (APC 1638 and COX2 -/-), as well as those to be developed in this proposal (Muc2 -/-, MUC2/beta-gal), will be used to test this hypothesis.